ARTIFICIAL INTELLIGENCE APPLICATIONS IN LASER SPECTROSCOPY DATA ANALYSIS Текст научной статьи по специальности «Медицинские технологии»

ARTIFICIAL INTELLIGENCE APPLICATIONS IN LASER SPECTROSCOPY DATA ANALYSIS

YURY V. KISTENEV,1 ALEXEY V. BORISOV,1 VLADIMIR V. PRISHEPA,1 VIKTOR E. SKIBA,1 AND IGOR

K. LEDNEV1,2

1 Tomsk State University, Tomsk 634050, Russian Federation 2 University at Albany, SUNY, Albany, NY 12222, USA

[email protected]

ABSTRACT

The report is devoted to applications of artificial intelligence (AI) methods for laser absorption and Raman spectroscopy data analysis, including data preprocessing, complex gas mixture decomposition, breath air analysis to detect a specific disease. Breath air analysis can be conducted through the chemical-composition-based and pattern-recognition-based approaches. AI methods like deep neural networks [1] can be combined with original chemometrics' methods like HAMAND [2] or (RSC) [3].

The principal issue in laser spectroscopy analysis of biological origin samples is associated with the latter decomposition, when the sample content is unknown fully. We suggested conducting this analysis by analyzing molecular compounds, which can be contained in the sample, by a "one-per-step" decomposition approach. Two methods of this idea implementation were created by us: HAMAND [2] or (RSC) [3]. HAMAND is based on extracting concentration of a target component by combining multivariate curve resolution with the addition method. RSC explores idea that a spectrum complexity is minimized when a definite component is exactly removed from it. We also plan to compare both methods with application artificial deep learning network application in the same task.

The research was carried out with the support of a grant under the Decree of the Government of the Russian Federation No. 220 of 09 April 2010 (Agreement No. 075-15-2021-615 of 04 June 2021) and Grant from the Ministry of Education and Science of Russia (Agreement No. 075-15-2021-1412 dated December 23, 2021, unique contract identifier RF— 2251 .62321X0012).

REFERENCES

[1] V. V. Prischepa, et al. Proc. SPIE 11582, 2020. P. 115821J; doi:10.1117/12.2581568

[2] M. Ando, I.K. Lednev, and H.-o Hamaguchi. In: Frontiers and Advances in Molecular Spectroscopy. 2018. P.369-378.

doi:10.1016/B978-0-12-811220-5.00011-3

[3] A.Borisov, et al. Journal of Breath Research 2021. P.027104. doi:10.1088/1752-7163/abebd4